Method, electronic device, and computer program product for verifying virtual avatar

Patent metadata

Field	Value
Publication number	US-12456181-B2
Application number	US-202318144285-A
Country	US
Kind code	B2
Filing date	May 8, 2023
Priority date	Apr 14, 2023
Publication date	Oct 28, 2025
Grant date	Oct 28, 2025

Abstract

Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for verifying a virtual avatar. The method includes: determining, based on the virtual avatar and a plurality of reference images in an image library, a first similarity set comprising a plurality of similarities between the virtual avatar and the plurality of reference images; and selecting a plurality of candidate images from the plurality of reference images based on the first similarity set and a predetermined number. The method further includes: determining a second similarity set comprising a plurality of similarities between the virtual avatar and the plurality of candidate images; and determining a verification result for the virtual avatar based on a subset in the first similarity set that is associated with the plurality of candidate images, and the second similarity set.

First claim

What is claimed is:

1. A method for verifying a virtual avatar, comprising:
determining, based on the virtual avatar and a plurality of reference images in an image library, a first similarity set comprising a plurality of similarities between the virtual avatar and the plurality of reference images;
selecting a plurality of candidate images from the plurality of reference images based on the first similarity set and a predetermined number;
determining a second similarity set comprising a plurality of similarities between the virtual avatar and the plurality of candidate images;
determining a verification result for the virtual avatar based on a subset in the first similarity set that is associated with the plurality of candidate images, and the second similarity set; and
generating a non-fungible token for the virtual avatar based on the verification result;
wherein determining a verification result for the virtual avatar comprises:
merging the subset and the second similarity set to obtain a third similarity set;
determining the verification result to be normal in response to absence of a similarity in the third similarity set that exceeds a threshold value; and
determining the verification result to be abnormal in response to presence of a similarity in the third similarity set that exceeds the threshold value.

2. The method according to claim 1 , wherein determining the first similarity set comprising the plurality of similarities between the virtual avatar and the plurality of reference images comprises:
extracting a facial region in the virtual avatar;
extracting target features of the facial region of the virtual avatar;
extracting a plurality of reference facial regions in the plurality of reference images; and
extracting a plurality of reference features of the plurality of reference facial regions of the plurality of reference images.

3. The method according to claim 2 , further comprising:
determining the plurality of similarities between the virtual avatar and the plurality of reference images based on the target features and the plurality of reference features;
ranking the plurality of similarities between the virtual avatar and the plurality of reference images;
selecting the predetermined number of similarities based on said ranking; and
wherein selecting a plurality of candidate images from the plurality of reference images based on the first similarity set and a predetermined number comprises:
determining a plurality of reference images corresponding to the selected predetermined number of similarities as the plurality of candidate images.

4. The method according to claim 3 , wherein selecting the predetermined number of similarities based on said ranking comprises:
dividing the plurality of reference features into a plurality of sub-vectors, wherein each sub-vector has a corresponding one of a plurality of sub-spaces;
generating corresponding sub-codebooks based on the plurality of sub-spaces;
mapping the plurality of sub-vectors to codewords in the corresponding sub-codebooks; and
determining the plurality of similarities between the virtual avatar and the plurality of reference images based on distances between the codewords and a plurality of codewords for the plurality of candidate images.

5. The method according to claim 2 , further comprising:
storing the target features and the plurality of reference features to a data storage apparatus.

6. The method according to claim 1 , wherein determining a second similarity set comprising a plurality of similarities between the virtual avatar and the plurality of candidate images comprises:
determining the plurality of similarities between the virtual avatar and the plurality of candidate images via a predetermined face matcher, wherein the face matcher has a matching algorithm different from a similarity algorithm for determining the first similarity set; and
determining the plurality of similarities between the virtual avatar and the plurality of candidate images as the second similarity set.

7. The method according to claim 1 , wherein merging the subset and the second similarity set to obtain a third similarity set comprises at least one of:
weighting the subset and the second similarity set based on predetermined weights;
searching the image library based on the subset to obtain the plurality of candidate images, and ranking the plurality of candidate images based on the second similarity set;
ranking the plurality of candidate images using the second similarity set; and
ranking the plurality of candidate images separately based on the subset and the second similarity set, and taking an intersection of the results of the two rankings.

8. The method according to claim 1 , further comprising:
generating, in response to the result of verifying the virtual avatar being normal, the non-fungible token for the virtual avatar for use as a virtual asset of a user; or
indicating to the user, in response to the result of verifying the virtual avatar being abnormal, a non-uniqueness of the virtual avatar.

9. The method according to claim 8 , further comprising:
generating metadata for the virtual asset in response to generating the non-fungible token for the virtual avatar for use as the virtual asset of the user; and
storing the metadata into a decentralized network.

10. An electronic device, comprising:
at least one processor; and
memory coupled to the at least one processor, wherein the memory has instructions stored therein which, when executed by the at least one processor, cause the electronic device to perform actions comprising:
determining, based on a virtual avatar and a plurality of reference images in an image library, a first similarity set comprising a plurality of similarities between the virtual avatar and the plurality of reference images;
selecting a plurality of candidate images from the plurality of reference images based on the first similarity set and a predetermined number;
determining a second similarity set comprising a plurality of similarities between the virtual avatar and the plurality of candidate images;
determining a verification result for the virtual avatar based on a subset in the first similarity set that is associated with the plurality of candidate images, and the second similarity set; and
generating a non-fungible token for the virtual avatar based on the verification result;
wherein determining a verification result for the virtual avatar comprises:
merging the subset and the second similarity set to obtain a third similarity set;
determining the verification result to be normal in response to absence of a similarity in the third similarity set that exceeds a threshold value; and
determining the verification result to be abnormal in response to presence of a similarity in the third similarity set that exceeds the threshold value.

11. The electronic device according to claim 10 , wherein determining a second similarity set comprising a plurality of similarities between the virtual avatar and the plurality of candidate images comprises:
extracting a facial region in the virtual avatar;
extracting target features of the facial region of the virtual avatar;
extracting a plurality of reference facial regions in the plurality of reference images; and
extracting a plurality of reference features of the plurality of reference facial regions of the plurality of reference images.

12. The electronic device according to claim 11 , wherein the actions further comprise:
determining the plurality of similarities between the virtual avatar and the plurality of reference images based on the target features and the pluralit

Assignees

• Dell Products Lp

Inventors

• Chen Si
• Wang Zijia
• Lin Zhenzhen

Classifications

• G06T2207/30201

  Face · CPC title

• G06Q30/018

  Certifying business or products · CPC title

• G06Q30/06

  Buying, selling or leasing transactions · CPC title

• G06Q40/04

  Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange · CPC title

• G06Q50/18

  Legal services · CPC title